University of Cambridge > Talks.cam > Isaac Newton Institute Seminar Series > Lecture 1: Some old and new results on Information-Based Complexity

Lecture 1: Some old and new results on Information-Based Complexity

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact info@newton.ac.uk.

ASC - Approximation, sampling and compression in data science

We give a short introduction to IBC and present some basic definitionsand a few results. The general question is: How many function values (or values of other functionals) of $f$ do we need to compute $S(f)$ up to an error $\epsilon$? Here $S(f)$ could be the integral or the maximum of $f$.
In particular we study the question: Which problems are tractable? When do we have the curse of dimension?  In the second talk we discuss complexity results for numerical integration. In particular we present results for the star discrepancy, the curse of dimension for $C^k$ functions, and results for randomized algorithms




This talk is part of the Isaac Newton Institute Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity